Compositions to Measure Protein Concentration
I. Field of the Invention
The present invention relates to the improvement of a previously described chemical composition which is used to measure protein concentration. The improvement relates to the discovery of an optimal composition which results in lower variance when measuring different proteins. In addition the assay results in a color change that can easily be detected by eye and that changes visibly and predictably with protein concentration. The discovery allows rapid one step protein concentration measurement for research, diagnostic, clinical and other biotechnology applications.
II. Discussio of the Background Art
Protein concentration measurement is a pre-requisite for many biotechnological applications. In particular the research field uses protein assays to accurately determine the concentration of a particular protein for quantitative analysis. The diagnostic and clinical field use assays to monitor bodily fluids for protein content, from which it can be determined whether the patient is responding to treatment or whether they have normal body fluid composition. Other applications of a protein assay is in the field of biotechnology where protein quantitation is crucial for reproducibility or monitoring processes.
The current methods of protein assay are deficient in one or more of the following parameters; speed, accuracy, sensitivity, protein to protein variability, linear range of measurement or simplicity of operation. Most currently used assays depend on one dye or t complex that generates a different color when binding to proteins. The most widely used research protein assay (Bradford's reagent (Bradford M.M. 1976, Anal. Biochem.72, 248-
254), varies widely in its color generating ability between different proteins. This reagent varies up to five fold when similar protein mass is used to compare different proteins (Read
l
and Northcote, Anal. Biochem. 116, 53-64. 1981; Marshall T. and Williams K. M. 1992, Anal. Biochem. 204, 107-109). However, the assay is simple and quick to perform. The Lowry assay (Lowry O.H., Rosebrough N.J., Farr A.L. and Randall R.J. 1951. J. Biol. Chem. 193, 265-275) also depends on a single chemical complex containing copper and it requires an extended sample preparation method before protein assay, so this method is not rapid. However, the Lowry assay does have low variability when comparing different proteins. Finally, the method of Fujita et al. (Fujita Y., Mori I. and Kitano S. 1983, Bunseki Kagaku, 32, E379-E386) which also relies on just one color generating complex (pyrogallol red and molybdate ions) has a useful range of measurement and low variability between proteins, but it lacks sensitivity at low protein concentrations. Protein measurements can also be obtained by direct absorbance measurement of a protein solution at 254-280nm, however this assay is requires a previous determination of an extinction coefficient and also many other substances absorb in this wavelength range, which interferes with the readings.
Prior to the invention disclosed herein, there remains a need a protein assay that combines the following properties; a) A protein assay that has less variability than Bradford method, b) A protein assay that is more rapid than the Lowry method, c) A protein assay that has a greater signal than the Fujita method. d) A protein assay with an easy to read method of quantitation.
HI, Summary of the invention
The present invention provides a novel chemical composition that rapidly measures protein concentration with good sensitivity and low variability. It has a simple one step procedure with high accuracy and a good range of measurement. Identifying specific high quality batches of Serva Blue G dye is a key factor in this invention as is the use of 5 to 6% phosphoric acid. Finally, the addition of an inactive dye that changes the initial color of the solution from rusty brown to emerald green creates a color change that can easily be detected by eye and that changes visibly and predictably with protein concentration. This allows protein concentration measurements without the use of a spectrophotometer. In combining these useful characters, the present invention creates a novel composition.
The protein assay described here measures protein at a concentration of 1.0- 5,000mg/L and in volumes as small as 5ul of sample. Samples can be determined in quantitatively in a spectrophotometer set to measure absorbance between 570 and 615nm. Alternatively, a color coded chart can be used to estimate protein concentration in a semi- quantitative manner, which has not been feasible before now because the Bradford reagent is not accurate enough between different proteins (so a different chart is required for each protein) and the Lowry and Fujita methods do not exhibit enough color change to support a visible inspection of protein dependent color change. The vessel of measurement can be a single tube, cuvette or other translucent vessel such as a 96- well or 384-well plate.
The present invention measures the concentration of five pure proteins (bovine serum albumin, soybean trypsin inhibitor, lysozyme, immunoglobulin-G and ovalbumin), to within a coefficient of variation of +/-10% of the mean. Whereas the same proteins are measured by the Bradford method to +/-87% of the mean. The same proteins are measured by the Lowry method to +/- 18% of the mean and by the Fujita method by +/- 19% of the mean. Thus the present invention improves upon the Bradford, Lowry and Fujita methods by reducing the variability between different proteins. In addition, the signal obtained with the present invention is much greater (approximately five fold more) than that of the Fujita method.
The present invention measures protein in urine with increased sensitivity compared to the Fujita method and much more rapidly than the Lowry method.
Simple methods of production, a long shelf life and ambient temperature shipping make this reagent suitable for a wide range of applications requiring reproducible, simple and low variance protein assays.
Chronology of the invention:
1999 Started to work on pyrogallol red (Fujita method) as a possible product (e.g. lab book 00042, p. 7,8,17,18,33,35,40,41,43,48,49,54,58,61-63)
2000 January discovered a one to one mix of Bradford and Fujita' s reagents gave a good invariability between proteins, but a small linear range (e.g. lab book 00028, p.84). 2000 February to April discovered that a modified Bradford reagent could give a good invariability between proteins on its own (e.g. lab book 00022, p.38-64).
2000 February to March designed the logo for the "Advanced protein assay reagent" and the label for the product (see label printer).
2000 May; optimized the invention formulation (e.g. lab book 00022, p. 81). 2000 May; added a red dye to achieve a reproducible green to blue color change instead of brown to blue (e.g. lab book 00022, p. 85).
2000 May; determined that different batches of Serva Blue G (the main color generating dye) gave different responses in the assay. In particular the character of batch lot#: 10766 was superior to batch lot#: 04183, when considering the extended linear range of the assay (e.g. lab book 00022, p. 81-83).
2000 May; created color charts that can be used to measure protein concentration in a semi- quantitative manner, and more rapidly than using a spectrophotometer (e.g. lab book 00022, P. 87).
2000 May; created production, packaging, storage and shipping methods for the commercial product (e.g. lab book 00022, p. 88).
IV. Description of the drawings
Figure 1 - Optimizing the dye content of the invention. The Serva blue G dye content was varied between 35 and 70mg/l, and BSA was kept at lOOug protein per ml of reagent. Absorbance was measured at 595nm.
Figure 2 - Optimizing the phosphoric acid content of the invention. Phosphoric acid content was varied between 5 and 8% and five different proteins were measured at lOug protein per ml of reagent.
Figure 3 - Identifying optimal batches of Serva Blue G dye. Different batches of Serva Blue G were compared using the optimal composition of the invention.
Figure 4 - Optimizing the visible color change of the reagent. Pyrogallol red was added to the optimal composition of protein assay reagent at between 3 and 12ug dye per ml of reagent. Protein was measured at lOug/ml and it was shown that above 6ug/ml of Pyrogallol red the dye reduces the protein induced color change. Emerald green quality was
determined visually and scored from 0 to 0.25 with the highest values creating the most clearly visible color change.
Figure 5a - Spectral analysis of Bradford reagent and the Invention in the absence or presence of protein helps quantify differences in their physical properties. Figure 5 a - lOOug BSA per ml of reagent was used to compare the reagents. Observe the difference in absorption at 450 and 640nm peaks in the absence of protein, this reflects the use of different components; phosphoric acid concentration, optimal Serva blue batches and the pyrogallol red dye as a color changing component. The samples containing protein also reflect this difference. Figure 5b - Bradford reagent plus and minus pyrogallol red dye, there is less difference between Bradford reagent and Bradford reagent plus pyrogallol red at 640nm area compared to the Invention, indicating that pyrogallol red changes mainly the 450nm peak which changes the color of the reagent without interfering with protein induced color changes in the 570 to 615nm area.
Figure 6 - Linear range of protein assay (Invention, Bradford, Lowry and Fujita reagents). BSA was used to compare the fore mentioned protein assays by measuring 0.25, 0.50, 1.0, 2.0, 5.0, 10, 20, 30, 40, 50, 60, 70 and lOOug of protein per ml of reagent.
Figure 7 - Variance of protein assay (Invention, Bradford, Lowry and Fujita reagents). Five different pure proteins (bovine serum albumin, soyabean trypsin inhibitor, lysozyme, immunoglobulin-G and ovalbumin) were gravimetrically measured to lg/L. Protein content was assayed with the Invention, Bradford, Lowiy and Fujita methods over a final concentration of 2 to 20mg/L. The mean values +/- standard deviations at 20mg/L are 0.59+A 0.059, 0.30+/-0.26, 0.20+/0.04 and 0.21+/-0.04 respectively.
Figure 8 - Determination of levels of interference by pure chemicals. Different chemicals commonly used in biological research were added at increasing concentration to the optimal composition. Highest concentrations of chemicals where there was no color change are presented as well as the concentrations that interfere with measuring lOug of BSA per ml reagent.
* - Color change is measured at OD595nm, for the non-interfering concentration shown here the OD595nm must change less than 0.05 OD units when lOul of chemical is added to l.Oml of reagent.
** - Protein concentration is lOug per ml reagent, color change is measured at OD595nm, for the non-interfering concentration shown here the OD595nm must change less than 0.05 OD units when lOul of chemical is added to l.Oml of reagent.
Figure 9 - Measurement of eleven urine samples to compare the Invention to Lowry and Fujita reagents. Five urine samples from one female and six samples from one male human were assayed for protein content. The respective absorbance values were plotted against each other to determine how similar measurements were. R2 values were 0.55 for Invention versus Fujita, 0.84 for Invention versus Lowry, and 0.45 for Lowry versus Fujita.
Figure 10 - Color coded chart for semi-quantitative protein measurement and the label of the product which reflects this color chart. Because the invention created less than 25% variation when measuring different proteins, the color change could be compared to a printed chart with colors representing 0 to 30ug protein per ml reagent. This figure shows such a chart in color, plus a table providing the optimal color composition in CMYK color code which are suitable for printing purposes.
V. Detailed description of the invention
The present invention provides an effective protein assay for general research, biotechnology and clinical/diagnostic fields. The protein assay is simple, rapid, accurate, has a low protein to protein variability, a good working range and good sensitivity. Its advantages over currently available methods are that it has lower variance than the Bradford method, it is simpler and more rapid than the Lowry method, and it has better sensitivity than the Fujita method. These characteristics are achieved by creating a formulation based on the Bradford reagent, and optimizing the components for desirable characteristics. In addition, we created a reagent that changes visibly and predictably from a green color to a blue color on addition of protein, this property allows the end user to obtain semi-quantitative protein concentrations using a color chart based on protein induced changes in the color. The color chart makes the assay very rapid and simple, and allows the end user to obtain semi-quantitative data even in the absence of a spectrophotometer which are expensive.
VI Development of the assay
The present invention emerged from a project aim to combine the useful properties of various protein assay reagents into an assay that was simple, rapid, accurate, sensitive, with a good range of measurement and a low variance between different proteins. The Bradford reagent performs well except for the variance between proteins which is a very poor property of this method. The Fujita's reagent performs well except for its sensitivity which is approximately five fold less than Bradford reagent. So it was likely that a combination of the reagents would create a novel reagent with the combined useful properties. Therefore, we performed mixing experiments to determine whether there was an optimal ratio of Bradford to Fujita reagents that retained useful characteristics of both reagents. However the pH that the Bradford reagent operates at 1.5 to 2,0 is too low for the operation of the Fujita reagent, pH 2.5 to 3.0. However, the dye (pyrogallol red) in the Fujita reagent changes the color of the Bradford reagent from a brown/bluey/green to emerald green. Although this dye is inactive it creates a color change that is clearly visible and varies predictably over a protein range of 0 to 30 ug per ml reagent. This allows the use of a color chart for semi-quantitative protein determination. This alerted us to the possibility of using multiple dyes in the same protein assay reagent to create a useful reagent with improved properties.
In order to optimize the visible color change characteristic of the reagent we titrated the pyrogallol red content whilst scoring for emerald green quality (Figure 4). We found that a point between 3.0 and 6.0 ug/ml pyrogallol red gave optimal green color whilst not interfering with the reaction between Serva blue G dye and protein. The spectral differences between the Invention solution and Bradford, both with and without BSA protein at lOOug ml reagent are shown in Figure 5a. Notice the lower peak at 450nm and the higher peak at 650nm in the Invention without protein compared to the Bradford reagent, both these are novel characters of the Invention solution. In addition it is noticed that at 590nm there is very little difference between Bradford and the Invention either with or without protein. Adding pyrogallol red dye to the Bradford reagent did not have the same effect as adding it to the Invention solution, and this is reflected in the spectral analysis, see Figure 5b. Thus the combination of optimized protein assay components in the Invention solution plus the pyrogallol red dye has created an identifiable spectral difference that results in improved visible color change characters.
The combination, of dyes has created a new reagent with properties that are superior to currently available colorimetric protein quantitation assays. The following properties are documented; a) The Bradford reagent's good linear range has been retained and its invariability has been improved, b) The Fujita reagent's dye has been utilized to create a final color change from green to blue rather than brown/bluey-green to blue (Bradford), or red to slightly purple/red (Fujita).
It is probable that other reagents containing multiple dye combinations will create different and possibly improved properties over this invention.
Via Determining the optimal composition of the protein assay reagent.
Serva blue G dye was varied over a range of 35 to 70mg/l. At low concentrations 35 to 50 mg/ml the maximum signal (measured with lOOug BSA per ml reagent) was lower (0.8 to 1.2 OD595nm) than the higher concentrations. However, at 60 to 70mg/l of dye, a precipitate formed after addition of the water. Thus the optimal concentration of Serva blue G was 55mg/l (Figure 1).
We tested the variability between five different proteins (bovine serum albumin, soybean trypsin inhibitor, lysozyme, immunoglobulin-G and ovalbumin) whose concentrations were gravimetrically determined. By varying the phosphoric acid content we found that as phosphoric acid content increased from 4% to 8% (v/v) the variability between the samples increased from approximately 20% to greater than 300% (Figure 2). The phosphoric acid has
two other functions in this assay, that is to keep the pH constant as a buffer, and to keep the dyes soluble. Therefore the optimal concentration of phosphoric acid was kept to 6% (v/v) because most of the variability occurred with higher concentrations.
We noticed that there was variation between batches of dye, when measuring BSA at lOOug protein per ml reagent. Batch lot# 10766 gave a mean maximum signal of 1.43, whereas batch lot# 04183 gave a value of 1.25 (Figure 3), thus it is important to screen batches and determine the one with the greatest signal generating capability which will create a product with a larger maximum range.
Vlb. Measuring different proteins.
Further studies compared seven different pure proteins. At 20ug protein per ml of reagent, the Bradford reagent gave a mean value of 0.30 with a coefficient of variation (cy) of 86%. Whereas the Fujita reagent gave a mean value 0,21 with a cv of 19%, and the Lowry reagent gave a mean of 0.20 with a cv of 18%. The invention gave a mean value of 0.56 with a cv of 13%. By retaining the high signal (with BSA) of the Bradford reagent and optimizing for invariability, the invention creates a higher mean signal with less variability between proteins (Figures 4 and 7).
Next we tested the linear range of protein assay by comparing the invention versus Bradford versus Lowry versus Fujita reagents. BSA was used to compare the fore mentioned protein assays by measuring 0.25, 0.50, 1.0, 2.0, 5.0, 10, 20, 30, 40, 50, 60, 70 and lOOug of protein per ml of reagent (see Figure 6). We noticed that the Fujita reagent was the most linear over the largest range of protein concentration (from 0 to 50ug protein per ml of reagent), this is because the low sensitivity which creates a lower slope to increasing concentrations of protein. The invention reagent retained a good linear range that is similar to Bradford reagent (0 to 40ug protein per ml reagent).
Vic. Measuring other chemical species for interference with the protein assay reagent.
The invention solution was mixed with various pure chemicals to see whether they reacted with the solution or whether they interfered with the reagents ability to detect and measure protein. As shown in Figure 8, there were nine classes of chemicals tested: these were divalent cations, metal chelating agents, detergents, organic solvents, reducing agents, antifoaming agents, buffers, protein denaturants and acids. Of these classes only the detergents
significantly affected the reagent or its reaction with proteins, all other chemicals were compatible to the invention solution. The level of detergents that were compatible with the invention solution were in the worldng range for most protein containing solutions. Using the methods described in V4, V5 or Examples 1,2 and 3 there is no problem with the reagent/detergent combinations at the concentrations shown in Figure 8.
V2 Preparation of the invention reagent
The Invention reagent has been optimized to create the following composition. Up to 40% variations in individual components can also create a useful reagent.
Invention reagent
For IL of IX add
55mg Serva blue G (lot: 10766 or similar)
4.5mg Pyrogallol red dye
50ml Ethanol (95% pure, 3 A grade or purer) (can be substituted for methanol)
60ml Phosphoric acid (85% or similar)
890ml nanopure water (<6mega ohm) For a 250L batch of 5X concentrate (final product formulation):
68.8 g Serva blue G (lot: 10766 or similar)
5.6 g Pyrogallol red dye
62.5 1 Ethanol (95% pure, 3 A grade or purer) (can be substituted for methanol)
75 1 Phosphoric acid (85% or similar)
1121 nanopure water (<6mega ohm)
Production process
The 2501 batch of 5X concentrate is made in a 300-5001 plastic drum container. The dyes, ethanol and phosphoric acid are added together and mixed vigorously for lh. The water is then added and mixed for a further 30min. Bottles (500ml) are then filled with the 5X reagent and then placed in the 4°C for 24h. After this period labels are placed on the bottles at 4°C, this keeps the labels from wrinkling. Bottles are then sealed with tamper resistant shrink wrap tubing.
V3 Storage of the invention solution
The 5X concentrated invention solution or the IX version can be stored between -70°C and 50°C, more preferably between 0°C and 30°C and most preferably at 4°C +/-4°C. The vessel of storage can be any hard plastic container or any container that is resistant to phosphoric acid (pHl.5-2.0) and ethanol/methanol combinations, in addition an air-tight sealed lid stops loss of volatile components.
V4 General method of use of the invention solution The present invention is used in a general protein assay format with the following procedure, 50ul of protein solution (containing 3.0 to 150ug protein) is pipetted into 5ml of IX strength invention solution. Alternatively the 5X invention solution can be added directly to four volumes of dilute protein solution in order to measure low protein amounts (e.g. 0.5 to 3.0ug/ml). In both cases, the absorbance is read at one wavelength between 500 and 700nm, or more preferably between 570 and 615nm and most preferably at 590nm +/-20nm. Using the equation:
Eq. 1 A - eel,
where A =* absorbance, e ~ extinction coefficient , c = concentration of protein in g/ml and 1 = pathlength of the light beam, the concentration of protein is determined. The mean extinction coefficient of the invention solution is approximately 33 ml mg'1 cm"1. Alternative methods of use are detailed in the Examples section.
V5 Diagnostic applications of the invention solution
Many diseases (e.g. Kidney disease) are indicated by the presence of increased levels of protein in bodily fluids (e.g. urine). The present invention is used in the diagnostic protein assay format with the following procedure, either lOul of protein solution (containing 3.0 to 150ug protein) is pipetted into 5ml of IX strength invention solution and the absorbance is read as described in V4 above, and Examples 1 to 3. Alternatively other arrangeme'nts may be made to automate the assay procedure whereby there may be a dilution step prior to the assay, for example with serum protein determination the initial level of protein concentration may be up to 60mg/ml, which is too high to be assayed without dilution. In Example 2 and Figure 9,
urine samples have been measured with the present invention and compared to a currently used clinical method based on the Fujita reagent and the Lowry reagent. There is a poor correlation between the Fujita and Invention procedures (R2 = 0.55) and a better correlation between the Lowry and Invention (R2 = 0.84). Thus the Invention reagent has approximately four times the sensitivity of the Fujita reagent and may be detecting different components of protein or interfering agents in urine when compared to the Fujita reagent.
V6 Clinical applications
The routine measurement of bodily fluids (urine, serum, spinal fluid etc.) for protein content can set individual values for comparison to possible disease states at later times. The present invention can be used in a clinical protein assay format with the following procedure, 3ul of protein solution (containing 0.05 to 15ug protein) is pipetted into 0.30ml of IX strength invention solution in a 96-well plate and the absorbance is measured as described in V4 and Example 1 with a plate reader or continuous monitoring absorbance measurement system.
V7 High through-put format
The present invention can be used in a high through-put protein assay format with the following procedure, 3.0 to lOul of protein solution (0.05 to 15ug protein) is pipetted into 0.3ml of IX strength invention solution that is pre-aliquoted into a 96-well plate (alternatively 384-well plates can be used with 1.0 to 3.0ul of protein (0.01 to 4ug protein) and 80ul of reagent, and the absorbance is measured as described in V4 and Example 1 with an automated plate reader set up.
Examples of the invention
Example 1: Research and Biotechnology applications
The present invention is used in a general protein assay format with the following procedure, 50ul of protein solution is pipetted into 5ml of IX strength invention solution. Alternatively the 5X invention solution can be added directly to four volumes of dilute protein solution in order to measure low protein amounts (e.g. 0,5 to lOOug/ml). In both cases, the absorbance is read at one wavelength between 500 and 700nm, or more preferably between 560 and 640nm and most preferably at 596nm +/-20nm. Using the equation:
Eq. 1 A = eel,
where A - absorbance, e = extinction coefficient , c = concentration of protein in mg/ml and 1 = pathlength of the light beam, the concentration of protein is determined. The mean extinction coefficient of the invention solution is approximately 33 ml mg"1 cm"1.
Example 2: Use in measuring protein in urine samples (diagnostic and clinical applications).
The present invention measures protein in urine with increased sensitivity compared to the Fujita method and much more rapidly than the Lowry method. As shown in Figure 9, eleven urine samples were measured for protein concentration using the invention solution and a currently used clinical reagent (Fujita's reagent). There was poor coorelation between methods r = 0.54, which may indicate that the Invention solution is detecting different protein species compared to the Fujita reagent..
Example 3: Use of the color coded chart for semi-quantitative protein assay.
A color coded chart can be used to estimate protein concentration in a semi- quantitative manner, which has not been feasible before now because the Bradford reagent is not accurate enough between different proteins (so a different chart would be required for each protein) and the Lowry and Fujita methods do not exhibit enough color change to support a visible inspection of protein dependent color change. The vessel of measurement
can be a single tube, cuvette or other translucent vessel such as a 96-well or 384-well plate. See Figure 10 for a representative chart, color codes for printing purposes and the bottle label that reflects the color change.